A new class of explanations for classifiers with non-binary features

DOI10.1007/978-3-031-43619-2_8MaRDI QIDQ6545524zbMATH OpenFDO

Authors Chunxi Ji, Adnan Darwiche

Publication date 29 May 2024

Full work available at URL https://doi.org/10.1007/978-3-031-43619-2_8

zbMATH Keywords

decision graphsexplainable AIprime implicants/implicates

Mathematics Subject Classification ID

Logic in artificial intelligence (68T27)

Recommendations

• Feature necessity \& relevancy in ML classifier explanations
• On the reasons behind decisions
• Non-monotonic explanation functions
• How to explain individual classification decisions
• A Symbolic Approach for Counterfactual Explanations

Cites work

• A New Algorithm for Generating Prime Implicants
• A unified logical framework for explanations in classifier systems
• An incremental method for generating prime implicants/implicates
• CNF and DNF considered harmful for computing prime implicants/implicates
• Consequence finding algorithms
• Explaining black-box classifiers: properties and functions
• Explanation in artificial intelligence: insights from the social sciences
• Graph-Based Algorithms for Boolean Function Manipulation
• scientific article; zbMATH DE number 5852793 (Why is no real title available?)
• scientific article; zbMATH DE number 3860199 (Why is no real title available?)
• scientific article; zbMATH DE number 1348482 (Why is no real title available?)
• scientific article; zbMATH DE number 1946872 (Why is no real title available?)
• On generating the irredundant conjunctive and disjunctive normal forms of monotone Boolean functions
• On quantifying literals in Boolean logic and its applications to explainable AI
• On Tackling Explanation Redundancy in Decision Trees
• On the reasons behind decisions
• Same-decision probability: a confidence measure for threshold-based decisions

Cited in

(9)

• Non-monotonic explanation functions
• Persuasive contrastive explanations for Bayesian networks
• A logic for binary classifiers and their explanation
• A unified logical framework for explanations in classifier systems
• Explanation of pseudo-Boolean functions using cooperative game theory and prime implicants
• Explaining robust classification through prime implicants
• Feature necessity \& relevancy in ML classifier explanations
• On the failings of Shapley values for explainability
• Explainability via short formulas: the case of propositional logic with implementation

This page was built for publication: A new class of explanations for classifiers with non-binary features

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6545524)